• is.na(),any(is.na()), all(is.na()),count(), and functions from naniar like gg_miss_var() and miss_var_summary can help determine if we have NA values
• miss_var_which() can help you drop columns that have any missing values.
• filter() automatically removes NA values
• drop_na() can help you remove NA values
• NA values can change your calculation results
• think about what NA values represent - don’t drop them if you shouldn’t
• replace_na() will replace `NA values with a particular value

• case_when() can recode entire values based on conditions
  • remember case_when() needs TRUE ~ variable to keep values that aren’t specified by conditions, otherwise will be NA
• stringr package has great functions for looking for specific parts of values especially filter() and str_detect() combined
  • also has other useful string manipulation functions like str_replace() and more!
  • separate() can split columns into additional columns
  • unite() can combine columns

📃Cheatsheet

In this module, we will show you how to:

1. Reshape data from wide to long
2. Reshape data from long to wide
3. Merge Data/Joins

Data is stored differently in the tibble.

Wide: has many columns

# A tibble: 1 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514           0.511

Long: column names become data

# A tibble: 3 × 3
  State   name            value
  <chr>   <chr>           <dbl>
1 Alabama June_vacc_rate  0.516
2 Alabama May_vacc_rate   0.514
3 Alabama April_vacc_rate 0.511

Wide: multiple columns per individual, values spread across multiple columns

# A tibble: 2 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514           0.511
2 Alaska           0.627         0.626           0.623

Long: multiple rows per observation, a single column contains the values

# A tibble: 6 × 3
  State   name            value
  <chr>   <chr>           <dbl>
1 Alabama June_vacc_rate  0.516
2 Alabama May_vacc_rate   0.514
3 Alabama April_vacc_rate 0.511
4 Alaska  June_vacc_rate  0.627
5 Alaska  May_vacc_rate   0.626
6 Alaska  April_vacc_rate 0.623

https://github.com/gadenbuie/tidyexplain/blob/main/images/tidyr-pivoting.gif

Wide: Easier for humans to read

# A tibble: 2 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514           0.511
2 Alaska           0.627         0.626           0.623

Long: Easier for R to make plots & do analysis

# A tibble: 6 × 3
  State   name            value
  <chr>   <chr>           <dbl>
1 Alabama June_vacc_rate  0.516
2 Alabama May_vacc_rate   0.514
3 Alabama April_vacc_rate 0.511
4 Alaska  June_vacc_rate  0.627
5 Alaska  May_vacc_rate   0.626
6 Alaska  April_vacc_rate 0.623

tidyr allows you to “tidy” your data. We will be talking about:

• pivot_longer - make multiple columns into variables, (wide to long)
• pivot_wider - make a variable into multiple columns, (long to wide)

The reshape command exists. Its arguments are considered more confusing, so we don’t recommend it.

You might see old functions gather and spread when googling. These are older iterations of pivot_longer and pivot_wider, respectively.

wide_vacc <- read_csv(
  file = "https://daseh.org/data/wide_vacc.csv")

wide_vacc

# A tibble: 1 × 3
  June_vacc_rate May_vacc_rate April_vacc_rate
           <dbl>         <dbl>           <dbl>
1          0.516         0.514           0.511

long_vacc <- wide_vacc %>% pivot_longer(cols = everything())
long_vacc

# A tibble: 3 × 2
  name            value
  <chr>           <dbl>
1 June_vacc_rate  0.516
2 May_vacc_rate   0.514
3 April_vacc_rate 0.511

wide_vacc

# A tibble: 1 × 3
  June_vacc_rate May_vacc_rate April_vacc_rate
           <dbl>         <dbl>           <dbl>
1          0.516         0.514           0.511

long_vacc <- wide_vacc %>% pivot_longer(cols = everything(),
                                        names_to = "Month",
                                        values_to = "Rate")
long_vacc

# A tibble: 3 × 2
  Month            Rate
  <chr>           <dbl>
1 June_vacc_rate  0.516
2 May_vacc_rate   0.514
3 April_vacc_rate 0.511

Newly created column names are enclosed in quotation marks.

https://daseh.org/data/CO_heat_er_visits_DenverBoulder_wide.csv

library(dasehr)
wide <- nitrate
head(nitrate)

# A tibble: 6 × 11
   year quarter pop_on_sampled_PWS `pop_0-3ug/L` `pop_>3-5ug/L` `pop_>5-10ug/L`
  <dbl> <chr>                <dbl>         <dbl>          <dbl>           <dbl>
1  1999 Q1                  106720         67775              0              32
2  1999 Q2                   85541         55476              0             212
3  1999 Q3                  559137        319252         231186             212
4  1999 Q4                   26995         25969            420               0
5  2000 Q1                   34793          5904              0              92
6  2000 Q2                  184521        157396              0              32
# ℹ 5 more variables: `pop_>10-20ug/L` <dbl>, `pop_>20ug/L` <dbl>,
#   `pop_on_PWS_with_non-detect` <dbl>, pop_exposed_to_exceedances <dbl>,
#   perc_pop_exposed_to_exceedances <dbl>

Let’s imagine we want to see what proportion of population exposed to different nitrate concentrations. Results should look something like:

# A tibble: 3 × 2
  conc_cat       avg_prop
  <chr>             <dbl>
1 pop_0-3ug/L    0.593   
2 pop_>10-20ug/L 0.000678
3 pop_>20ug/L    0.000129

long <- wide %>%
  select(!ends_with("exceedances")) %>% 
  pivot_longer(!c(year, quarter, pop_on_sampled_PWS),
               names_to = "conc_cat",
               values_to = "conc_count")
long

# A tibble: 528 × 5
    year quarter pop_on_sampled_PWS conc_cat                   conc_count
   <dbl> <chr>                <dbl> <chr>                           <dbl>
 1  1999 Q1                  106720 pop_0-3ug/L                     67775
 2  1999 Q1                  106720 pop_>3-5ug/L                        0
 3  1999 Q1                  106720 pop_>5-10ug/L                      32
 4  1999 Q1                  106720 pop_>10-20ug/L                      0
 5  1999 Q1                  106720 pop_>20ug/L                         0
 6  1999 Q1                  106720 pop_on_PWS_with_non-detect      38913
 7  1999 Q2                   85541 pop_0-3ug/L                     55476
 8  1999 Q2                   85541 pop_>3-5ug/L                        0
 9  1999 Q2                   85541 pop_>5-10ug/L                     212
10  1999 Q2                   85541 pop_>10-20ug/L                     60
# ℹ 518 more rows

Un-pivoted columns (year, quarter, pop_on_sampled_PWS) are similar

long

# A tibble: 528 × 5
    year quarter pop_on_sampled_PWS conc_cat                   conc_count
   <dbl> <chr>                <dbl> <chr>                           <dbl>
 1  1999 Q1                  106720 pop_0-3ug/L                     67775
 2  1999 Q1                  106720 pop_>3-5ug/L                        0
 3  1999 Q1                  106720 pop_>5-10ug/L                      32
 4  1999 Q1                  106720 pop_>10-20ug/L                      0
 5  1999 Q1                  106720 pop_>20ug/L                         0
 6  1999 Q1                  106720 pop_on_PWS_with_non-detect      38913
 7  1999 Q2                   85541 pop_0-3ug/L                     55476
 8  1999 Q2                   85541 pop_>3-5ug/L                        0
 9  1999 Q2                   85541 pop_>5-10ug/L                     212
10  1999 Q2                   85541 pop_>10-20ug/L                     60
# ℹ 518 more rows

Let’s make the conc_count into a proportion.

long <- long %>%
  mutate(conc_prop = conc_count / pop_on_sampled_PWS)
long

# A tibble: 528 × 6
    year quarter pop_on_sampled_PWS conc_cat                conc_count conc_prop
   <dbl> <chr>                <dbl> <chr>                        <dbl>     <dbl>
 1  1999 Q1                  106720 pop_0-3ug/L                  67775  0.635   
 2  1999 Q1                  106720 pop_>3-5ug/L                     0  0       
 3  1999 Q1                  106720 pop_>5-10ug/L                   32  0.000300
 4  1999 Q1                  106720 pop_>10-20ug/L                   0  0       
 5  1999 Q1                  106720 pop_>20ug/L                      0  0       
 6  1999 Q1                  106720 pop_on_PWS_with_non-de…      38913  0.365   
 7  1999 Q2                   85541 pop_0-3ug/L                  55476  0.649   
 8  1999 Q2                   85541 pop_>3-5ug/L                     0  0       
 9  1999 Q2                   85541 pop_>5-10ug/L                  212  0.00248 
10  1999 Q2                   85541 pop_>10-20ug/L                  60  0.000701
# ℹ 518 more rows

Now our data is more tidy, and we can take the averages easily!

long %>% 
  group_by(conc_cat) %>% 
  summarize("avg_prop" = mean(conc_prop))

# A tibble: 6 × 2
  conc_cat                   avg_prop
  <chr>                         <dbl>
1 pop_0-3ug/L                0.593   
2 pop_>10-20ug/L             0.000678
3 pop_>20ug/L                0.000129
4 pop_>3-5ug/L               0.182   
5 pop_>5-10ug/L              0.0189  
6 pop_on_PWS_with_non-detect 0.206   

There are many ways to select the columns we want. Check out https://dplyr.tidyverse.org/reference/dplyr_tidy_select.html to look at more column selection options.

pivot_wider() - spreads row data into columns (tidyr package)

• names_from = the old column whose contents will be spread into multiple new column names.
• values_from = the old column whose contents will fill in the values of those new columns.

{wide_data} <- {long_data} %>% 
  pivot_wider(names_from = {Old column name: contains new column names},
              values_from = {Old column name: contains new cell values})

long_vacc

# A tibble: 3 × 2
  Month            Rate
  <chr>           <dbl>
1 June_vacc_rate  0.516
2 May_vacc_rate   0.514
3 April_vacc_rate 0.511

wide_vacc <- long_vacc %>% pivot_wider(names_from = "Month", 
                                       values_from = "Rate") 
wide_vacc

# A tibble: 1 × 3
  June_vacc_rate May_vacc_rate April_vacc_rate
           <dbl>         <dbl>           <dbl>
1          0.516         0.514           0.511

long

# A tibble: 528 × 6
    year quarter pop_on_sampled_PWS conc_cat                conc_count conc_prop
   <dbl> <chr>                <dbl> <chr>                        <dbl>     <dbl>
 1  1999 Q1                  106720 pop_0-3ug/L                  67775  0.635   
 2  1999 Q1                  106720 pop_>3-5ug/L                     0  0       
 3  1999 Q1                  106720 pop_>5-10ug/L                   32  0.000300
 4  1999 Q1                  106720 pop_>10-20ug/L                   0  0       
 5  1999 Q1                  106720 pop_>20ug/L                      0  0       
 6  1999 Q1                  106720 pop_on_PWS_with_non-de…      38913  0.365   
 7  1999 Q2                   85541 pop_0-3ug/L                  55476  0.649   
 8  1999 Q2                   85541 pop_>3-5ug/L                     0  0       
 9  1999 Q2                   85541 pop_>5-10ug/L                  212  0.00248 
10  1999 Q2                   85541 pop_>10-20ug/L                  60  0.000701
# ℹ 518 more rows

wide <- long %>%
  select(!c(pop_on_sampled_PWS, conc_count)) %>%
  pivot_wider(names_from = "quarter", values_from = "conc_prop")
wide

# A tibble: 132 × 6
    year conc_cat                         Q1       Q2       Q3      Q4
   <dbl> <chr>                         <dbl>    <dbl>    <dbl>   <dbl>
 1  1999 pop_0-3ug/L                0.635    0.649    0.571    0.962  
 2  1999 pop_>3-5ug/L               0        0        0.413    0.0156 
 3  1999 pop_>5-10ug/L              0.000300 0.00248  0.000379 0      
 4  1999 pop_>10-20ug/L             0        0.000701 0        0      
 5  1999 pop_>20ug/L                0        0        0        0      
 6  1999 pop_on_PWS_with_non-detect 0.365    0.348    0.0152   0.0224 
 7  2000 pop_0-3ug/L                0.170    0.853    0.485    0.881  
 8  2000 pop_>3-5ug/L               0        0        0.00820  0.00244
 9  2000 pop_>5-10ug/L              0.00264  0.000173 0        0.00101
10  2000 pop_>10-20ug/L             0        0        0        0      
# ℹ 122 more rows

Reshaping can be helpful for your assessment of two group_by() categories.

long %>% 
  group_by(conc_cat, quarter) %>% 
  summarize("avg_prop" = mean(conc_prop)) %>% 
  pivot_wider(names_from = "quarter", values_from = "avg_prop")

`summarise()` has grouped output by 'conc_cat'. You can override using the
`.groups` argument.

# A tibble: 6 × 5
# Groups:   conc_cat [6]
  conc_cat                         Q1        Q2        Q3       Q4
  <chr>                         <dbl>     <dbl>     <dbl>    <dbl>
1 pop_0-3ug/L                0.389    0.655     0.717     0.610   
2 pop_>10-20ug/L             0.00100  0.000562  0.000407  0.000741
3 pop_>20ug/L                0.000325 0.0000481 0.0000409 0.000102
4 pop_>3-5ug/L               0.321    0.149     0.0903    0.168   
5 pop_>5-10ug/L              0.0255   0.0131    0.00984   0.0271  
6 pop_on_PWS_with_non-detect 0.263    0.182     0.183     0.194   

• tidyr package helps us convert between wide and long data
• pivot_longer() goes from wide -> long
  • Specify columns you want to pivot
  • Specify names_to = and values_to = for custom naming
• pivot_wider() goes from long -> wide
  • Specify names_from = and values_from =

🏠 Class Website

💻 Lab

“Combining datasets”

• Merging/joining data sets together - usually on key variables, usually “id”
• ?join - see different types of joining for dplyr
• inner_join(x, y) - only rows that match for x and y are kept
• full_join(x, y) - all rows of x and y are kept
• left_join(x, y) - all rows of x are kept even if not merged with y
• right_join(x, y) - all rows of y are kept even if not merged with x
• anti_join(x, y) - all rows from x not in y keeping just columns from x.

data_As <- read_csv(
  file = "https://daseh.org/data/data_As_1.csv")
data_cold <- read_csv(
  file = "https://daseh.org/data/data_cold_1.csv")

data_As

# A tibble: 2 × 3
  State   June_vacc_rate May_vacc_rate
  <chr>            <dbl>         <dbl>
1 Alabama          0.516         0.514
2 Alaska           0.627         0.626

data_cold

# A tibble: 2 × 2
  State  April_vacc_rate
  <chr>            <dbl>
1 Maine            0.795
2 Alaska           0.623

https://github.com/gadenbuie/tidyexplain/blob/main/images/inner-join.gif

ij <- inner_join(data_As, data_cold)

Joining with `by = join_by(State)`

ij

# A tibble: 1 × 4
  State  June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>           <dbl>         <dbl>           <dbl>
1 Alaska          0.627         0.626           0.623

https://raw.githubusercontent.com/gadenbuie/tidyexplain/main/images/left-join.gif

“Everything to the left of the comma”

lj <- left_join(data_As, data_cold)

Joining with `by = join_by(State)`

lj

# A tibble: 2 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514          NA    
2 Alaska           0.627         0.626           0.623

# install.packages("tidylog")
library(tidylog)
left_join(data_As, data_cold)

Joining with `by = join_by(State)`
left_join: added one column (April_vacc_rate)
> rows only in x 1
> rows only in y (1)
> matched rows 1
> ===
> rows total 2

# A tibble: 2 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514          NA    
2 Alaska           0.627         0.626           0.623

https://raw.githubusercontent.com/gadenbuie/tidyexplain/main/images/right-join.gif

“Everything to the right of the comma”

rj <- right_join(data_As, data_cold)

Joining with `by = join_by(State)`
right_join: added one column (April_vacc_rate)
> rows only in x (1)
> rows only in y 1
> matched rows 1
> ===
> rows total 2

rj

# A tibble: 2 × 4
  State  June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>           <dbl>         <dbl>           <dbl>
1 Alaska          0.627         0.626           0.623
2 Maine          NA            NA               0.795

lj2 <- left_join(data_cold, data_As)

Joining with `by = join_by(State)`
left_join: added 2 columns (June_vacc_rate, May_vacc_rate)
> rows only in x 1
> rows only in y (1)
> matched rows 1
> ===
> rows total 2

lj2

# A tibble: 2 × 4
  State  April_vacc_rate June_vacc_rate May_vacc_rate
  <chr>            <dbl>          <dbl>         <dbl>
1 Maine            0.795         NA            NA    
2 Alaska           0.623          0.627         0.626

https://raw.githubusercontent.com/gadenbuie/tidyexplain/main/images/full-join.gif

fj <- full_join(data_As, data_cold)

Joining with `by = join_by(State)`
full_join: added one column (April_vacc_rate)
> rows only in x 1
> rows only in y 1
> matched rows 1
> ===
> rows total 3

fj

# A tibble: 3 × 4
  State   June_vacc_rate May_vacc_rate April_vacc_rate
  <chr>            <dbl>         <dbl>           <dbl>
1 Alabama          0.516         0.514          NA    
2 Alaska           0.627         0.626           0.623
3 Maine           NA            NA               0.795

data_As <- read_csv(
  file = "https://daseh.org/data/data_As_2.csv")
data_cold <- read_csv(
  file = "https://daseh.org/data/data_cold_2.csv")

data_As

# A tibble: 2 × 2
  State   state_bird      
  <chr>   <chr>           
1 Alabama wild turkey     
2 Alaska  willow ptarmigan

data_cold

# A tibble: 3 × 3
  State  vacc_rate month
  <chr>      <dbl> <chr>
1 Maine      0.795 April
2 Alaska     0.623 April
3 Alaska     0.626 May  

lj <- left_join(data_As, data_cold)

Joining with `by = join_by(State)`
left_join: added 2 columns (vacc_rate, month)
> rows only in x 1
> rows only in y (1)
> matched rows 2 (includes duplicates)
> ===
> rows total 3

Data including the joining column (“State”) has been duplicated.

lj

# A tibble: 3 × 4
  State   state_bird       vacc_rate month
  <chr>   <chr>                <dbl> <chr>
1 Alabama wild turkey         NA     <NA> 
2 Alaska  willow ptarmigan     0.623 April
3 Alaska  willow ptarmigan     0.626 May  

Note that “Alaska willow ptarmigan” appears twice.

https://github.com/gadenbuie/tidyexplain/blob/main/images/left-join-extra.gif

unloadNamespace() does the opposite of library().

unloadNamespace("tidylog")

By default joins use the intersection of column names. If by is specified, it uses that.

full_join(data_As, data_cold, by = "State")

# A tibble: 4 × 4
  State   state_bird       vacc_rate month
  <chr>   <chr>                <dbl> <chr>
1 Alabama wild turkey         NA     <NA> 
2 Alaska  willow ptarmigan     0.623 April
3 Alaska  willow ptarmigan     0.626 May  
4 Maine   <NA>                 0.795 April

You can join based on multiple columns by using something like by = c(col1, col2).

If the datasets have two different names for the same data, use:

full_join(x, y, by = c("a" = "b"))

Entries in data_As but not in data_cold

anti_join(data_As, data_cold, by = "State")

# A tibble: 1 × 2
  State   state_bird 
  <chr>   <chr>      
1 Alabama wild turkey

Entries in data_cold but not in data_As

anti_join(data_cold, data_As, by = "State") # order switched

# A tibble: 1 × 3
  State vacc_rate month
  <chr>     <dbl> <chr>
1 Maine     0.795 April

• Merging/joining data sets together - assumes all column names that overlap
  • use the by = c("a" = "b") if they differ
• inner_join(x, y) - only rows that match for x and y are kept
• full_join(x, y) - all rows of x and y are kept
• left_join(x, y) - all rows of x are kept even if not merged with y
• right_join(x, y) - all rows of y are kept even if not merged with x
• Use the tidylog package for a detailed summary
• antijoin(x, y) shows what is only in x (missing from y)

🏠 Class Website

💻 Lab

Image by Gerd Altmann from Pixabay

We might want to determine what indexes ARE in the first dataset that AREN’T in the second.

For this to work, the datasets need the same columns.

We’ll just select the index using select().

A_states <- data_As %>% select(State)
cold_states <- data_cold %>% select(State)

States in A_states but not in cold_states

dplyr::setdiff(A_states, cold_states)

# A tibble: 1 × 1
  State  
  <chr>  
1 Alabama

States in cold_states but not in A_states

dplyr::setdiff(cold_states, A_states)

# A tibble: 1 × 1
  State
  <chr>
1 Maine

Why did we use dplyr::setdiff?

There is a base R function, also called setdiff that requires vectors.

In other words, we use dplyr:: to be specific about the package we want to use.

More set operations can be found here: https://dplyr.tidyverse.org/reference/setops.html

https://sebkrantz.github.io/collapse/

Might be helpful if your data is very large. However, dplyr and tidyr functions are great for most applications.